Research On Internal Model PID Control Of Sewage Treatment Based On LSSVM Method

The shortage of freshwater resources and the increasingly serious water pollution remind us of the necessity of sewage treatment.Biological Aerated Filter?BAF?,as a typical process of wastewater treatment,has the characteristics of non-linearity,strong coupling and large lag.These characteristics make it difficult to establish its accurate mathematical model,and there are many external disturbances in the field.Traditional control methods are difficult to control its effluent index effectively.In this paper,Biological Aerated Filter system is taken as the research object.First of all,the COD and effluent NH₃-N concentrations which are the important water quality parameters,are chosen as the output of the system in quite a few outputs,and the Statistical Product and Service Solutions software is used to choose the influence on the selected output based on the principal component analysis method.The largest input,as the input to the study,finally selected the influent COD and the influent NH3-N concentration as input to the system,thereby simplifying the Biological Aerated Filter system into a two-input dual-output system.Secondly,on the basis of studying the theory of SVM,LSSVM is proposed to recognize the simplified Biological Aerated Filter system,and Genetic Algorithm is used to optimize the LSSVM.The model of the simplified Biological Aerated Filter system is obtained.The simulation results reveal that the model recognized by GA-LSSVM can be used as the model of the simplified Biological Aerated Filter system.After that,because of the coupling characteristics of the Biological Aerated Filter system,the LSSVM direct inverse model method is used to construct the?order inverse system of the BAF system?LSSVM is also optimized by Genetic Algorithm?,and the?order inverse system is associated in series with the original system to form a pseudo-linear system,thereby getting rid of the coupling characteristics of the Biological Aerated Filter system.Finally,Internal model PID control is introduced to improve the anti-interference ability and robustness of the control system,and corresponding simulation experiments are carried out.The concrete content includes inverse model identification experiment,decoupling experiment and system.The control experiment can be seen from the simulation results of the experiment.The control strategy designed in this paper has good tracking performance and strong anti-interference ability,and has good applicability.The main contributions of this paper are as follows:the Biological Aerated Filter system is simplified by SPSS software based on principal component analysis,and the system is identified and decoupled by LSSVM improved by genetic algorithm.At the same time,the internal model PID control scheme is designed,and the feasibility of this scheme is verified by simulation.